based on the idea of https://github.com/dla-marbach/indexer the go indexer can be used to extract metadata from files while speeding up the whole process of the identification cascade.
Validation of Mediafiles with ffmpeg and imagemagick is supported too.
go get github.com/je4/indexer
go build github.com/je4/indexer/cmd/identify
identify -cfg indexer.toml
curl -X POST --data-binary @query.json http://localhost:81
query.json:
{
"url": "https://upload.wikimedia.org/wikipedia/commons/thumb/5/54/Stift_Melk_Nordseite_01.jpg/750px-Stift_Melk_Nordseite_01.jpg",
"actions": ["siegfried","identify","ffprobe","tika"],
"forcedownload": "^image/.*$",
"headersize": 5000
}
- url: mandatory field (file:///...)
- actions: optional field, list of identifiers to use
- forcedownload: optional field, regexp of mimetypes, which should be downloaded completely
- headersize: optional field, size of header which is downloaded for format recognition
Copyright 2020 Jürgen Enge, info-age GmbH, Basel
Licensed under the Apache License, Version 2.0